lg-mako: black screen

could you try downgrading mesa and libdrm to the versions available from alpine's edge repository?

By foldedcascode on 2019-03-11T01:47:49

or, might also be worth seeing what happens if you rebuild the device package for armv7, looks like it's still listed as armhf

By foldedcascode on 2019-03-11T01:51:28

Same result using armv7.
Added dmesg.lg-mako.armv7.log

By gouchi on 2019-03-11T21:17:48

Is this like that bullhead

By rahmanshaber on 2019-03-12T04:12:52

Yes indeed it looks like similar.

Actually I succeed to boot to Weston using :

pmbootstrap install --no-fde

By gouchi on 2019-03-13T21:53:59

mentioned in commit wiki@cf8e3a9d

By postmarketOS Wiki bot on 2019-03-14T07:42:52

This happens on hammerhead, too. #221 (closed)

By tallero on 2019-03-24T21:43:12

I don't believe this is the same issue as above, but I will followup here since this issue is open and its title matches my case.

My mako boots to a black screen. In my case, it is weston that fails to start.

In dmesg I have

[ 1749.601348] elogind-daemon[3189]: Failed to determine whether /sys/fs/cgroup is a mount point: Symbolic link loop
[ 1749.602386] elogind-daemon[3189]: Failed to mount cgroup at /sys/fs/cgroup/elogind: No such file or directory
[ 1749.602447] elogind-daemon[3189]: Failed to allocate manager object: Symbolic link loop

and logread shows

Oct  9 16:28:06 lg-mako daemon.info supervise-daemon[3175]: Supervisor command line: supervise-daemon lightdm --start /usr/bin/lightdm -- 
Oct  9 16:28:06 lg-mako daemon.info supervise-daemon[3177]: Child command line: /usr/bin/lightdm 
Oct  9 16:28:06 lg-mako daemon.info dbus-daemon[1240]: [system] Activating service name='org.freedesktop.login1' requested by ':1.54' (uid=0 pid=3177 comm="/usr/bin/lightdm " label="kernel") (using servicehelper)
Oct  9 16:28:31 lg-mako auth.notice dbus-daemon[1240]: [system] Failed to activate service 'org.freedesktop.login1': timed out (service_start_timeout=25000ms)
Oct  9 16:28:31 lg-mako authpriv.info : pam_unix(lightdm-autologin:session): session opened for user user by (uid=0)
Oct  9 16:28:31 lg-mako daemon.info dbus-daemon[1240]: [system] Activating service name='org.freedesktop.login1' requested by ':1.56' (uid=0 pid=3197 comm="lightdm --session-child 11 14 " label="kernel") (using servicehelper)
Oct  9 16:28:56 lg-mako auth.notice dbus-daemon[1240]: [system] Failed to activate service 'org.freedesktop.login1': timed out (service_start_timeout=25000ms)
Oct  9 16:28:56 lg-mako authpriv.err : pam_elogind(lightdm-autologin:session): Failed to create session: Operation timed out
Oct  9 16:28:57 lg-mako user.notice user:weston: Date: 2019-10-09 -03
Oct  9 16:28:57 lg-mako user.notice user:weston: [16:28:57.011] weston 7.0.0
Oct  9 16:28:57 lg-mako user.notice user:weston:                https://wayland.freedesktop.org
Oct  9 16:28:57 lg-mako user.notice user:weston:                Bug reports to: https://gitlab.freedesktop.org/wayland/weston/issues/
Oct  9 16:28:57 lg-mako user.notice user:weston:                Build: v20190809-679-g4a31b325d3
Oct  9 16:28:57 lg-mako user.notice user:weston: [16:28:57.011] Command line: weston -c /etc/xdg/weston/weston.ini
Oct  9 16:28:57 lg-mako user.notice user:weston: [16:28:57.011] OS: Linux, 3.4.0-lineageos, #11-postmarketOS SMP PREEMPT Mon Dec 3 21:41:15 UTC 2018, armv7l
Oct  9 16:28:57 lg-mako user.notice user:weston: [16:28:57.011] Using config file '/etc/xdg/weston/weston.ini'
Oct  9 16:28:57 lg-mako user.notice user:weston: [16:28:57.012] Output repaint window is 7 ms maximum.
Oct  9 16:28:57 lg-mako user.notice user:weston: [16:28:57.012] Loading module '/usr/lib/libweston-7/fbdev-backend.so'
Oct  9 16:28:57 lg-mako user.notice user:weston: [16:28:57.015] initializing fbdev backend
Oct  9 16:28:57 lg-mako user.notice user:weston: [16:28:57.017] logind: not running in a systemd session
Oct  9 16:28:57 lg-mako user.notice user:weston: [16:28:57.017] logind: cannot setup systemd-logind helper (-61), using legacy fallback
Oct  9 16:28:57 lg-mako user.notice user:weston: [16:28:57.018] fatal: fbdev backend should be run using weston-launch binary, or your system should provide the logind D-Bus API.
Oct  9 16:28:57 lg-mako user.notice user:weston: [16:28:57.018] fatal: failed to create compositor backend
Oct  9 16:28:57 lg-mako authpriv.info : pam_unix(lightdm-autologin:session): session closed for user user
Oct  9 16:28:57 lg-mako daemon.warn supervise-daemon[3176]: /usr/bin/lightdm, pid 3177, exited with return code 1
Oct  9 16:28:57 lg-mako daemon.info supervise-daemon[3218]: Child command line: /usr/bin/lightdm 
Oct  9 16:28:57 lg-mako daemon.info dbus-daemon[1240]: [system] Activating service name='org.freedesktop.login1' requested by ':1.57' (uid=0 pid=3218 comm="/usr/bin/lightdm " label="kernel") (using servicehelper)

A workaround is to start weston as root using

rc-service lightdm stop
export XDG_RUNTIME_DIR=/tmp
weston -B fbdev-backend.so --tty=1 --no-config&
weston-terminal&

By pedro on 2019-10-09T19:43:06

I can reproduce the issue, just tried to boot lg-mako to test a merge request and it bootest straight to blackscreen, even with pmaports.git master. I've tried xfce4 and weston.

Directly afterwards, I've tried weston with the samsung-i9100 and that worked. So it seems to be a lg-mako specific issue.

Unfortunately I don't have any time to debug this (working on postmarketOS infrastructure... and going through MR/issues backlog), but at least here's an additional data point. Thanks to everybody who's looking into it.

By Oliver Smith on 2019-10-13T19:05:42

mentioned in commit wiki@62692f1b

By postmarketOS Wiki bot on 2019-10-13T19:43:05

I tried a fresh installation using xfce4 today. mako still boots to a black screen. Here follows the full Xorg.logXorg.0.log, where X crashes with a segfault:

[   138.401] (EE) Segmentation fault at address 0x4
[   138.401] (EE) 
Fatal server error:
[   138.401] (EE) Caught signal 11 (Segmentation fault). Server aborting
[   138.402] (EE) 
[   138.402] (EE) 
Please consult the The X.Org Foundation support 
         at http://wiki.x.org
 for help. 
[   138.402] (EE) Please also check the log file at "/var/log/Xorg.0.log" for additional information.
[   138.402] (EE) 
[   138.661] (EE) Server terminated with error (1). Closing log file.

I have no clue as to where to start. Anyone has a working X running on mako and can post a full Xorg.log for comparison?

By pedro on 2019-10-24T19:09:18

lightdm fails to start xfce4 session, and as there is no greater installed, a black screen remains. If lightdm-gtk-greater is installed, it shows up after a while.

Still no clues as to why xfce4 session does not start though:

[+0.05s] DEBUG: Launching process 3880: /usr/bin/X :0 -seat seat0 -auth /var/run/lightdm/root/:0 -nolisten tcp vt7 -novtswitch
[+0.05s] DEBUG: XServer 0: Waiting for ready signal from X server :0
...
[+1.73s] DEBUG: Got signal 10 (SIGUSR1) from process 3880
...
[+1.80s] DEBUG: Session pid=3889: Running command /usr/bin/lightdm-session startxfce4
...
[+1.81s] DEBUG: Activating VT 7
[+11.09s] DEBUG: Process 3880 terminated with signal 6 (SIGABRT)
[+11.09s] DEBUG: XServer 0: X server stopped
[+11.09s] DEBUG: Releasing VT 7
...
[+11.09s] DEBUG: Seat seat0: Active display server stopped, starting greeter
[+11.09s] DEBUG: Seat seat0: Creating greeter session
[+11.09s] DEBUG: Seat seat0: Failed to find session configuration lightdm-gtk-greeter
[+11.09s] DEBUG: Seat seat0: Stopping; failed to start a greeter
...

lg-mako:~# cat /home/user/.xsession-errors
sh: a: unknown operand
<stdin>:0: warning: Unknown encoding: C.UTF-8
    gpg-agent[4329]: WARNING: "--write-env-file" is an obsolete option - it has no effect
gpg-agent: a gpg-agent is already running - not starting a new one

(xfce4-session:4293): xfce4-session-WARNING **: 16:50:30.528: gpg-agent returned no PID in the variables

(xfce4-session:4293): xfce4-session-WARNING **: 16:50:30.542: xfsm_manager_load_session: Something wrong with /home/user/.cache/sessions/xfce4-session-lg-mako:0, Does it exist? Permissions issue?
xfsettingsd: No window manager registered on screen 0.

(xfsettingsd:4336): xfsettingsd-WARNING **: 16:50:37.374: Failed to get the _NET_NUMBER_OF_DESKTOPS property.
xfce4-panel: No window manager registered on screen 0. To start the panel without this check, run with --disable-wm-check.
Gdk-Message: 16:50:38.299: xfsettingsd: Fatal IO error 11 (Resource temporarily unavailable) on X server :0.
.

Gdk-Message: 16:50:38.301: xfce4-panel: Fatal IO error 0 (No error information) on X server :0.

lightdm.log.old

By pedro on 2019-10-24T20:11:30

mentioned in merge request !1011 (merged)

By TAKIZAWA Fumiya on 2020-02-29T15:21:50

I spent a bit of time playing around, and this is what I have so far.

I picked the i3wm UI at pmbootstrap init time. As with everyone else in this issue, the display turns black after flashing the postmarketOS logo for a second, and after that the backlight turns on and off a few times but there's no other sign of activity.

I can ssh into the device, and from there see that Xorg is started, along with the i3 tools, but eventually dies.

Once logged in via ssh, I can run Xorg :0 and in one screen window and then run simple applications such as xeyes and xterm successfully: they show up on the phone's display, and I can move the cursor around with my finger. I can even run simple WMs such as Openbox and Window Maker without issues - though everything has a blue tint, reminiscent of the red tint bug that IIUC affected the device a while ago.

However, as soon as I try to run i3 or spectrwm, Xorg crashes.

When I try running Xorg under gdb, after installing musl-dbg and xorg-server-dbg, this is the result:

lg-mako:~$ sudo gdb --args Xorg :0
GNU gdb (GDB) 9.1
Copyright (C) 2020 Free Software Foundation, Inc.
License GPLv3+: GNU GPL version 3 or later <http://gnu.org/licenses/gpl.html>
This is free software: you are free to change and redistribute it.
There is NO WARRANTY, to the extent permitted by law.
Type "show copying" and "show warranty" for details.
This GDB was configured as "armv6-alpine-linux-musleabihf".
Type "show configuration" for configuration details.
For bug reporting instructions, please see:
<http://www.gnu.org/software/gdb/bugs/>.
Find the GDB manual and other documentation resources online at:
    <http://www.gnu.org/software/gdb/documentation/>.

For help, type "help".
Type "apropos word" to search for commands related to "word"...
Reading symbols from X...
Reading symbols from /usr/lib/debug//usr/bin/Xorg.debug...
(gdb) b CreatePicture
Breakpoint 1 at 0x104854: file picture.c, line 751.
(gdb) r
Starting program: /usr/bin/Xorg :0

X.Org X Server 1.20.7
X Protocol Version 11, Revision 0
Build Operating System: Linux 4.19.78-1-vanilla armv8l Alpine Linux
Current Operating System: Linux lg-mako 3.4.0-lineageos #11-postmarketOS SMP PREEMPT Tue Nov 5 05:56:49 UTC 2019 armv7l
Kernel command line:  uart_console=enable lcd_maker_id=primary lge.hreset=off lge.reset=mode_reset gpt=enable lge.kcal=0|0|0|x lge.rev=rev_11 androidboot.emmc=true androidboot.serialno=02207948d357d5d8 androidboot.bootloader=MAKOZ30f androidboot.baseband=mdm
Build Date: 14 January 2020  04:18:43PM

Current version of pixman: 0.38.4
        Before reporting problems, check http://wiki.x.org
        to make sure that you have the latest version.
Markers: (--) probed, (**) from config file, (==) default setting,
        (++) from command line, (!!) notice, (II) informational,
        (WW) warning, (EE) error, (NI) not implemented, (??) unknown.
(==) Log file: "/var/log/Xorg.0.log", Time: Fri Feb 28 02:22:27 2020
(==) Using system config directory "/usr/share/X11/xorg.conf.d"
[New LWP 2206]
[New LWP 2207]
[New LWP 2208]
[New LWP 2209]
[New LWP 2210]
[New LWP 2211]
[New LWP 2212]
[New LWP 2213]
[Detaching after fork from child process 2214]
The XKEYBOARD keymap compiler (xkbcomp) reports:
> Internal error:   Could not resolve keysym XF86FullScreen
Errors from xkbcomp are not fatal to the X server
[New LWP 2215]

Thread 1 "X" hit Breakpoint 1, CreatePicture (pid=2097166, pDrawable=0xb72362c0, pFormat=0xb6f59e20, vmask=1024, vlist=vlist@entry=0xb7209510, client=client@entry=0xb71edb90,
    error=0xbeef8cd8, error@entry=0xbeef8cd0) at picture.c:751
751     picture.c: No such file or directory.
(gdb) disp pFormat
1: pFormat = (PictFormatPtr) 0xb6f59e20
(gdb) c
Continuing.

Thread 1 "X" hit Breakpoint 1, CreatePicture (pid=2097185, pDrawable=0xb7236ed0, pFormat=0xb6f59e20, vmask=384, vlist=vlist@entry=0xb7209528, client=client@entry=0xb71edb90,
    error=0xbeef8cd8, error@entry=0xbeef8cd0) at picture.c:751
751     in picture.c
1: pFormat = (PictFormatPtr) 0xb6f59e20
(gdb) c
Continuing.

Thread 1 "X" hit Breakpoint 1, CreatePicture (pid=2097185, pDrawable=0xb7236a90, pFormat=0xb6f59e20, vmask=1024, vlist=vlist@entry=0xb7209510, client=client@entry=0xb71edb90,
    error=0xbeef8cd8, error@entry=0xbeef8cd0) at picture.c:751
751     in picture.c
1: pFormat = (PictFormatPtr) 0xb6f59e20
(gdb) c
Continuing.

Thread 1 "X" hit Breakpoint 1, CreatePicture (pid=2097188, pDrawable=0xb7237ca0, pFormat=0xb6f59e20, vmask=384, vlist=vlist@entry=0xb7209528, client=client@entry=0xb71edb90,
    error=0xbeef8cd8, error@entry=0xbeef8cd0) at picture.c:751
751     in picture.c
1: pFormat = (PictFormatPtr) 0xb6f59e20
(gdb) c
Continuing.

Thread 1 "X" hit Breakpoint 1, CreatePicture (pid=2097188, pDrawable=0xb727b480, pFormat=0xb6f59e20, vmask=1024, vlist=vlist@entry=0xb7209520, client=client@entry=0xb71edb90,
    error=0xbeef8cd8, error@entry=0xbeef8cd0) at picture.c:751
751     in picture.c
1: pFormat = (PictFormatPtr) 0xb6f59e20

Thread 1 "X" hit Breakpoint 1, CreatePicture (pid=2097212, pDrawable=0xb727b880, pFormat=0xb6f59e20, vmask=384, vlist=vlist@entry=0xb7209538, client=client@entry=0xb71edb90,
    error=0xbeef8cd8, error@entry=0xbeef8cd0) at picture.c:751
751     in picture.c
1: pFormat = (PictFormatPtr) 0xb6f59e20
(gdb) c
Continuing.

Thread 1 "X" hit Breakpoint 1, CreatePicture (pid=pid@entry=0, pDrawable=pDrawable@entry=0xb6fc9970, pFormat=0x0, pFormat@entry=0xbeef8c3c, vmask=vmask@entry=256,
    vlist=0xbeef8c1c, vlist@entry=0xbeef8c14, client=0xb6943e70, error=error@entry=0xbeef8c20) at picture.c:751
751     in picture.c
1: pFormat = (PictFormatPtr) 0x0
(gdb) c
Continuing.

Thread 1 "X" received signal SIGSEGV, Segmentation fault.
0xb6e5f8d0 in CreatePicture (pid=pid@entry=0, pDrawable=pDrawable@entry=0xb6fc9970, pFormat=0x0, pFormat@entry=0xbeef8c3c, vmask=vmask@entry=256, vlist=0xbeef8c1c,
    vlist@entry=0xbeef8c14, client=0xb6943e70, error=error@entry=0xbeef8c20) at picture.c:763
763     in picture.c
1: pFormat = (PictFormatPtr) 0x0
(gdb) bt
#0  0xb6e5f8d0 in CreatePicture (pid=pid@entry=0, pDrawable=pDrawable@entry=0xb6fc9970, pFormat=0x0, pFormat@entry=0xbeef8c3c, vmask=vmask@entry=256, vlist=0xbeef8c1c,
    vlist@entry=0xbeef8c14, client=0xb6943e70, error=error@entry=0xbeef8c20) at picture.c:763
#1  0xb6e203e8 in compNewPixmap (pWin=pWin@entry=0xb727b480, x=490, y=-1225371280, w=304, h=18) at compalloc.c:567
#2  0xb6e21168 in compAllocPixmap (pWin=pWin@entry=0xb727b480) at compalloc.c:604
#3  0xb6e1f7e0 in compCheckRedirect (pWin=pWin@entry=0xb727b480) at compwindow.c:167
#4  0xb6e1f894 in compRealizeWindow (pWin=0xb727b480) at compwindow.c:267
#5  0xb6dc7d78 in RealizeTree (pWin=pWin@entry=0xb727b480) at window.c:2617
#6  0xb6dca9f4 in MapWindow (client=0xb71edb90, pWin=0xb727b480) at window.c:2694
#7  MapWindow (pWin=0xb727b480, client=client@entry=0xb71edb90) at window.c:2665
#8  0xb6d9b714 in ProcMapWindow (client=0xb71edb90) at dispatch.c:845
#9  0xb6da2030 in Dispatch () at dispatch.c:478
#10 0xb6da5da0 in dix_main (argc=2, argv=0xbeef8db4, envp=<optimized out>) at main.c:276
#11 0xb6cdf7a8 in libc_start_main_stage2 (main=0xb6d8fe04 <main>, argc=2, argv=0xbeef8db4) at src/env/__libc_start_main.c:94
#12 0xb6d8fe78 in _start_c (p=<optimized out>) at crt/crt1.c:18
#13 0xb6d8fe28 in _start () at render.c:2065
Backtrace stopped: previous frame identical to this frame (corrupt stack?)
(gdb)

Looking at the CreatePicture() function in Xorg, the cause of the issue is pretty clear:

PicturePtr
CreatePicture(Picture pid,
              DrawablePtr pDrawable,
              PictFormatPtr pFormat,
              Mask vmask, XID *vlist, ClientPtr client, int *error)
{
    PicturePtr pPicture;
    PictureScreenPtr ps = GetPictureScreen(pDrawable->pScreen);

    pPicture = dixAllocateScreenObjectWithPrivates(pDrawable->pScreen,
                                                   PictureRec, PRIVATE_PICTURE);
    if (!pPicture) {
        *error = BadAlloc;
        return 0;
    }

    pPicture->id = pid;
    pPicture->pDrawable = pDrawable;
    pPicture->pFormat = pFormat;

    /* This is where the crash happens */
    pPicture->format = pFormat->format | (pDrawable->bitsPerPixel << 24);

So when pFormat is NULL, as is the case in the last iteration, dereferencing it will result in a crash.

Now, why you'd end up in a situation where the pointer is NULL in the first place... I have no idea O:-)

Interestingly, gdb displays different values for pFormat and pFormat@entry, suggesting that the value was changed after the function has been called, but that doesn't make a whole lot of sense to me: when setting the breakpoint to the function and displaying the value immediately, it's already NULL.

Another thing to note is that the whole "corrupt stack" thing might be a red herring, because it shows up in the backtrace even for previous iterations, where the function works correctly and doesn't result in a crash.

By Andrea Bolognani on 2020-03-14T18:33:23

Currently effort is switched to move to mainline Linux kernel, black screen is still issue (on -mainline), but hopefully get solved soon.

Updates here: https://wiki.postmarketos.org/wiki/Google_Nexus_4_(lg-mako)

By David Heidelberg on 2021-09-10T00:18:54

added No maintainer help wanted labels

lg-mako: black screen

Child items 0

Activity